RT3 Patch tasks (long, make sure you read end)

[milo]

Here's the list of things to try to fix in RT3. Stuff in brackets is my guess as to feasibility, the resources we'll need to implement that particular fix, and the amount of time it'll take.

Skill legend:

• A- Artist (ability to construct and texture 2D and 3D objects, preferably using Blender for 3D)
• C- Coder (ability to write and debug code in x86 assembly language)
• F- File Hacker (ability to modify RT3 resource files such as *.LCO and *.BCA)
• I- Interpreter (ability to translate text for inclusion in the RT3.LNG file, as well as release notes)
• M- Mapmaker (ability to create new RT3 maps, including rivers and ground cover)
• S- Scenario Writer (ability to write and debug RT3 scenario events)
• T- Tester (ability to note bugs and behavior changes when playing RT3 scenarios

Time legend:

• Hours- 2-48 hours of work
• Days- 2-14 days of work
• Weeks- 2-11 weeks of work
• Months- 2-12 months of work
• Years- it'd be faster to write a new game

THE TASKS

AI:

• improve AI's track-laying algorithm (CJNyfalt, milo) [30%,C,weeks] - May require scripted AI
• improve AI's business sense (Grandma Ruth) [30%,C,days] - May require scripted AI
• give player option to freeze AI after company 51% owned (ExpressWorld) [20%,C,days]
• optionally specify goals for AI in scenario (milo) [20%,C,weeks] - Requires scripted AI
• let AI take over/sabotage other companies (milo) [30%,C,days] - May require scripted AI
• fix AI's inability to recover from damaged track (milo) [30%,C,weeks]
• be able to write AI's algorithm using scripting language (milo) [30%,C,weeks] - Requires nearly same work as scripted events; may slow down game

Assets:

• add specialized stations, or at least cargo-controlled stations (JSS) [50%,ACF,days]
• add cargo types (canis39) [80%,AC?F,days]
• add new maps (canis39, Grandma Ruth) [100%,M,days]
• add new locomotives (canis39) [100%,AF,days]
• add new characters (not just portraits) (ExpressWorld) [80%,CI,days]
• improve building animations (Grandma Ruth) [50%,A,weeks]
• be able to supply patches and assets with scenarios (Gwizz) [30%,C,weeks] - Great idea! Tricky to implement, though - can't patch the executable or add assets while it's running, so it'd take a Civ IV-like restart-after-mod approach.

Bugfixes:

• fix color washout problem on multiple saves (Grandma Ruth) [40%,C,days]
• fix crashes when deleting track (milo) [60%,C,weeks]
• fix early exits from multiple-factor event loops (milo) [90%,C,hours]
• fix tests for deliveries to/from cities (milo) [50%,C,days]
• fix unimplemented events, such as specialized track counters (milo) [50%,C,days]
• prevent AIs from reusing logos (milo, davey917) [30%,C,hours] - Is it only the 'keystone' that gets reused?
• fix event display in debug window for long scenarios (milo) [30%,C,days]
• fix alignment of custom logos in message lines (davey917) [90%,C,hours]
• fix instabilities after deactivating players (milo) [80%,C,days]

Configuration:

• lower system requirements (Grandma Ruth) [10%,C,months] - If anything, patch will probably require a faster computer - sorry
• be able to turn off planes, non-evented storms, and other incidental fripperies (Grandma Ruth, davey917) [30%,C,weeks]

Editor:

• add copy+paste to events (JSS, Grandma Ruth, milo) [20%,C,weeks] - May have to settle for event export/import
• be able to delete territories (Grandma Ruth) [40%,C,days]

Events:

• add math expressions to events (JSS, Wolverine@MSU, Grandma Ruth, davey917, milo) [30%,C,days] - Requires script support
• add tests for installed assets (JSS) [40%,C,hours]
• event land/water deformation (milo) [40%,C,days]
• event particle source and town creation/destruction (milo) [40%,C,days]
• support event libraries (managers, train robberies, no maintenance, etc) (milo) [30%,C,days] - Requires script support
• support Python extensions to events (milo) [30%,C,months] - Hard, but opens up so many possibilities…
• output subsets of game state to a logfile (Wolverine@MSU) [40%,C,days] - Requires script support

Gameplay:

• add industry-only AI companies (JSS) [30%,C,days] - Free if we split up tracklaying rights from industry rights
• be able to run railroad without investing in industry (JSS, CJNyfalt) [90%,C?S,hours] - Probably fixable in scenario - disable industry build/buy, cut price of rails and engines
• add personally-owned industries (JSS, davey917) [20%,C,weeks] - Requires a whole new page in ledger, among other things
• be able to ship at a loss (JSS, dj, Grandma Ruth) [90%,C,hours] - Not sure about side-effects, though
• add option to disable 'any freight'/'any express' consists (dj) [30%,C,days]
• merge maintenance buildings into stations (CJNyfalt, Grandma Ruth, Steve) [70%,C?S,hours] - Fixable in scenario; might need coding to stop AIs building them
• add management overhead related to number of stations (CJNyfalt) [90%,S,hours] - Fixable in scenario
• split track-laying rights from train-running or industry-buying/building (HighVoltage, davey917) [30%,C,weeks] - Requires several new events and possibly more scenario file storage
• reduce industry profits relative to railroad profits (CJNyfalt, Orange46) [90%,S,hours] - Fixable in scenario
• add event for single train annual/lifetime best speed (EPH, milo) [30%,C,weeks] - May slow game down considerably
• add city-to-city or terr-to-terr load delivery tests (Wolverine@MSU) [20%,C,weeks] - Not sure if cargos have individual IDs; may slow down game
• be able to ship cargoes through intervening stations (Grandma Ruth) [80%,C,days] - Probably get this for free when adding 'ship at loss'
• split territories by geographical/eventable rather than visible/invisible (Grandma Ruth) [20%,C,weeks] - I think this might break more scenarios than it would fix
• be able to transfer company/personal wealth through campaign scenarios (Grandma Ruth) [20%,C,days] - Doable… but frankly the problem's often too >much< money, not too little
• disable train arrival notices above certain rate threshold (but permit overriding for 'special' trains) (davey917) [30%,C,days] - Disabling notices is not hard, overriding for special trains is much harder
• add custom newspaper heralds and sidebar headlines (davey917) [30%,AC?F,days] - Not sure how these are stored - but a nifty idea
• support drop-shipping: leave cargo at station for pickup by another train (hawk) [80%,A?CF?,days] - a) Suppress economics around depot b) Construct cargo 'holder' building
• be able to 'reserve' certain amount of cargo in each building (Steve) [60%,CF?,days] - I think this is also one possible implementation for drop-shipping
• support multiple-engine and double-ended trains (ExpressWorld) [50%,A?CF?,days] - Bombardiere already demonstrated one way to do part of this
• add multiple track types: narrow-gauge, maglev, monorail, etc. (ExpressWorld) [10%,ACFS,months] - Requires major changes to game engine
• add signals (ExpressWorld) [10%,ACFS,years] - Requires major changes to game engine
• increase effect of slow delivery on perishable cargos (ExpressWorld) [90%,F,hours]
• bring back rate wars from RT1 (ExpressWorld) [60%,CS,weeks] - more complicated than I thought, but still almost entirely doable through events
• bring back train robberies from RT2 (ExpressWorld) [50%,AC,days] - Probably add as alternative to train crash
• force out stockholders by paying premium (milo) [30%,C,days]
• support express traffic to/from warehouses and ports (milo) [90%,C,hours]
• reduce cargo natural flow rate (Steve) [60%,C,days]
• support 'all except' consists (Steve) [40%,C,weeks]
• add ability to create campaigns (Wolverine@MSU) [40%,CMS,weeks] - Amount of work depends on whether campaigns need videos and 'rail museum' support

Graphics:

• fix railbed and rail-connection graphics (JSS, dj) [70%,AC,days] - Lower bed, maybe fix joint mitering, maybe redraw railbed texture
• avoid showing that all structures are hollow 2D shells (JSS) [30%,C,days] - This probably requires a powerful graphics card… sure you want it?
• add camera that follows direction of train (milo) [50%,C,days] - Would probably have to replace train-locked cam, unfortunately
• slightly increase possible zoom distance to cover huge maps (milo) [80%,C,hours]

Rail Layout:

• prevent buildings from overlapping rails (JSS) [60%,C,hours] - Cost: harder to place track in towns or around stations
• cut down track 'exclusivity' radius to permit stuff like parallel rails (canis39) [80%,C,hours] - Cost: have to have finer touch to make connections to existing rails
• add cuts and fills when building rail through mountains (dj, milo) [40%,C,days] - Cost: may open some terraforming exploits
• lower bridges and shorten ramps (dj, CJNyfalt) [80%,C,hours] - Cost: overpasses may be too low if they use same code as bridges
• cut down river nonbuildable radius (CJNyfalt) [30%,C,days] - Radius is 1 square; not sure I can cut this without permitting building >in< river
• add manual option for bridges, tunnels, and junctions (bombardiere, Grandma Ruth) [30%,C,weeks] - This needs to be fleshed out more
• cut out terraforming effects under existing rails (ExpressWorld) [60%,C,days] - Cost: new buildings near rails potentially half-buried or suspended in air
• support bridge upgrades (ExpressWorld) [30%,C,weeks]

Tools:

• add tool to import 3D models from cheap editors (Blender?) (bombardiere) [70%,C,days]
• add tool to import and export heightmaps (?) [90%,C,days] - Halfway done - will finish this before I start patch
• add tool to generate realistic random maps and/or simplify DEM import (milo) [70%,C,weeks]
• make it trivial to submit scenario reviews (EPH) [70%,?,days]
• add tool to import and export events (milo) [70%,C,days]

Clearly we can't implement everything on this list - there's just way too much stuff. I'm thinking we should shoot for two separate patches:

• 1.06 - contains just quick fixes, to be released within two months; perhaps includes an updated Loco Pack as well
• 1.07 - includes a scripting engine and some set of harder fixes, to be released by the end of 2007 if ever; this is as close to RT4 as we're likely to get

So... it's time to vote on what we should do. I can't figure out a good way to use the polling mechanism to resolve this, so instead, I want you to post a message in this thread in which you list your top ten desired fixes from the list above. Also, if you can donate time to the project, please PM me with a list of the skills you can provide.

edit 2006/11/26: add ideas from Steve, Wolverine@MSU; correct rate wars estimate
edit 2006/11/28: add idea from Gwizz; add bugfix for player deactivation

[JayEff]

Milo, you have a longish pm.
<----- I,M,S,T


From your wee To Do list, the ones that I like the most are:
- Stronger AI. All of that list. Face it, RT3 is not good for multi player.
- Import and exportable events. Fine if you can copy paste. I make mistakes copying by eye from my planning spreadsheets.
- Turn off non-vented storms. I am working on a scenario with a complex array of weather events. I don't want a rain storm happening in the middle of my drought.
- Add math expressions to events. I really miss that from RT2.

[JSS]

Wow, this is some list.
I would suggest that to start out with you should consider items that have a 50% or better probability attached to it only.

So here would be my priority:
1.) Complete all Bug fixes.
fix crashes when deleting track (milo) [60%,C,weeks]
fix early exits from multiple-factor event loops (milo) [90%,C,hours]
fix tests for deliveries to/from cities (milo) [50%,C,days]
fix unimplemented events, such as specialized track counters (milo) [50%,C,days]
fix alignment of custom logos in message lines (davey917) [90%,C,hours]

2.) Graphic issues
fix railbed and rail-connection graphics (JSS, dj) [70%,AC,days] - Lower bed, maybe fix joint mitering, maybe redraw railbed texture

3.) Gameplay.
be able to ship at a loss (JSS, dj, Grandma Ruth) [90%,C,hours] - Not sure about side-effects, though

Notes:
be able to run railroad without investing in industry (JSS, CJNyfalt) [90%,C?S,hours] - Probably fixable in scenario - disable industry build/buy, cut price of rails and engines
Would the scenario disable not also prevent A.I. companies from owning industries? There should be the possibility of industry only A.I. companies.

add personally-owned industries (JSS, davey917) [20%,C,weeks] - Requires a whole new page in ledger, among other things
If I can start an Industry only A.I. company that I can switch back and forth from the RR company that would be fine. So the challenge will be to find a way to trick the game into accepting being the chairman of those two companies at the same time regardless of the amount of your share ownership.

Bring back rate wars from RT1 (ExpressWorld) [90%,S.days]
Are you certain that your time frame of days is sufficient? This was quite a process. Keeping track of all shipments as well as what was required to be delivered would be quite an undertaking no?

BTW, if you do need some help I am available. Let me state my qualifications:
Artist – Forget it.
Coder – Not in my lifetime (or yours).
File Hacker – Not if you want anything useful left.
Interpreter – What?
Mapmaker – Lets talk.
Scenario Writer – With enough other volunteers to troubleshoot this might work.
Tester – Ok, but I want you to know this is a real sacrifice on my part.

[milo]

Good questions, JSS!

On industry/RR mix: While it isn't possible to disable industry build/buy per-company, it is possible to hike the cost of such to ridiculous levels, or alternatively make RR building ruinous for a given company. Is the AI smart enough to recognize that spending $100M for one segment of rail is a bad thing? Unfortunately there's nothing to block one type of company from purchasing the other. See the 'independent territorial rights' project for a cleaner approach, but it takes quite a bit more effort.

I like your idea about switching chairmanships between two companies, but I'd argue that 'personal' ownership of industries means that the industry-only company should be wholly owned by you. It may be possible to permit starting new companies without switching chairmanship. You'd still have to manage a majority ownership in your original company before you could safely switch to the new company, of course.

I probably misunderstood rate wars - I never played RT1. How did they work? Testing which company delivered more is feasible under the current event system, though it could be easier; throwing the loser out of town is also feasible with territory-scope destruction, though right now it almost invariably crashes the game.

[JSS]

Milo, this is a brief description of how the rate wars worked. If you need it in more detail I will copy the pages from the Official Guide to Railroad Tycoon and make them available to you.

Rate wars can be used to acquire a desirable station belonging to a competitor. However, they’re best used as a weapon to weaken an opponent financially.
The golden rule of Rate Wars is to win them fast. This can-not be overemphasized. A long, drawn-out Rate War will put a severe drain on your finances. If you decide to start a Rate War, plan carefully so you can hit hard and fast; long, drawn-out conflicts are losing propositions.

Meet the City Council.
The city council vote at the end of the end of the first complete fiscal period is potentially determinative. The first vote (that does not count) occurs at the end of the fiscal period in which the Rate War was started. The next vote will be after a full fiscal period and can end the war if enough votes are gathered by one side.
The voting that occurs at the end of each fiscal period is based on deliveries in that period only. Deliveries in earlier fiscal periods are not included in the determination. Each railroad starts each fiscal period of a Rate War with zero tons delivered; zero tons picked up and zero votes.

How votes are cast.
You get votes for both cargo delivered and cargo picked up. The way the votes are cast differs for deliveries and pickups.
For each commodity that a Rate War city wants, there are two votes. You’ll get a vote if you deliver any amount of the commodity the city wants. If you deliver the commodity and the other railroad does not, you get two votes. Conversely, it gets two votes if it delivers a commodity which you do not.
Although it is impractical, ideally you would deliver one car carrying one ton of each commodity in demand. Remember, income is cut in half during a Rate War, so any additional effort to deliver more than you need is wasted.
Picking up cargo is different. Picking up is measured by tonnage. The more tonnage you haul away, the more likely you are to get votes.
One, two, or three votes are given for commodities hauled away. The number of votes depends on the total amount of tonnage delivered by both sides during the fiscal period. If the total tonnage is less than 40 tons, there is one vote. Two votes are produced by 40 – 79 tons. Eighty tons or greater produce three votes.
If there is but one vote available, the railroad that carries the most tonnage of that cargo gets the vote. If two votes are available, the line that carries twice as much as the competition gets both votes; if neither line has double what the other line has picked up each line gets a vote. Where there are three votes, the line that carries away the most tons get two votes. If that line carries more than twice as much as the other line, it gets all three votes.

[Elite Marksman]

My priorities would be the bug-fixes, and then the AI fixes. The AI in RT3 is a joke...

As for skills,
unless you want modern art, forget it.
zero coding skills.
no file hacking.
Fairly good German and passable Spanish
Never got around to finishing my map...
Not much of a scenario designer
I can test, but between school, bowling, and work, not a whole lot of time

[canis39]

Wow. Milo, that's an impressive list!

I don't know about a "top 10" for me...I agree with JSS that it probably makes the most sense to focus on the easiest changes. Maybe start with all the 80% and up for the first patch, and then do as much as possible between 50% and 80% for the "final" patch. Or something like that.

Sadly, I have no skills whatsoever. I would volunteer to be a tester, though! :)

[davey917]

milo:

lower bridges and shorten ramps (dj, CJNyfalt) [80%,C,hours] - Cost: overpasses may be too low if they use same code as bridges

I wonder, does anyone actually use overpasses?
They seem to be more for looks than any actual function since trains treat an 'at grade' crossing as if it were an overpass. I usually try not to let one occur, especially if I plan to build a station next to a competitor's.

From your list, I agree with many others, aim for the bugfixes first, and anything which can be done quickly or easily.

Would you want additional ideas here, or in the other thread? Everytime I think about it, I remember more bugs :( or wishes .

almost forgot: Thanks for the efforts!

[milo]

Would you want additional ideas here, or in the other thread? Everytime I think about it, I remember more bugs :( or wishes .

Post them in the other thread or PM me, and I'll update the list here. Thanks!

[bombardiere]

Great list Milo :D . However there are things which may have negative side effects. For example, I would leave the track bed alone, because it has an important grade effect. (the game builds embankments to help grades)

and yes, I use overpasses. Love them in fact, even if they have no real function. May be best thing is narrow the rivers.(take away extra reserved cell. )

Milo I have a personal request. We talked about this a time ago. I asked that can engines be "attached" with cargo files. (to create Multiple units and railcars) You said no, as the game does not search for *.cgo files for engines. Now I would like to ask if it is possible to add this to code. (assuming it doesn't mean loss of game performance)

F, half-A, and possibly M

[Grandma Ruth]

Thanks, Milo, what a hell of a job you're doing.
I agree we should go for the most possible fixes first.
I could have a go at writing scenarios, not very experienced but willing!
Also I'll test with pleasure. Only thing is I don't have that much time - just a few hours in the course of a week so you wouldn't want to give me anything with a short deadline to it.

[Steve]

milo, somethings might be doable like, prevent AI from purchasing industry in territory they have no rights.

Can you change the general cargo system to not flow more than 0.1 carload per cell per month? This would well up the cargo at the source somewhat so to make it a bit like RT2 where small stations would be built to overlap the mine or farm. In this case 1.2 loads would float away each year but 1.0 load would pile up at the source.

A much more difficult change would be to have a prevent mask on either a station owned by a player or the train consist. This would prevent the loading at the station of certain cargoes. Thus you could set the consist to all cargo but prevent iron or coal from being loaded and your steel mill's supplies stolen and taken to another station.

[Steve]

add copy+paste to events (JSS, Grandma Ruth, milo) [20%,C,weeks] - May have to settle for event export/import

You can do this now can't you? I think I remember spell checking the text in Trans-Siberian and pasting it in the box.

[JayEff]

You can copy paste text now, and this is really nice for using event planning documents or spreadsheets, making exact quotes of historical headlines, or using unusual characters not supported by the editor. For instance this is how I get French accents into my dialogs and city names. But I think what milo means is the ability to import and export the events in their entirety, perhaps as a csv or something similar.

[JSS]

Actually, I was not aware that one can because I tried it and it did not work for me. I guess I will have to try it again.

[Steve]

Actually, I was not aware that one can because I tried it and it did not work for me. I guess I will have to try it again.

You have to do it all in one go. You can't paste half a page and the try to paste add another page on to that. You put it all in your editor and then paste it all at once.

[Wolverine@MSU]

Here are 2 sets of my top-10. The first are the "easy" ones, the second are the "harder" ones. They are listed in order of preference:

Easy:
1- be able to ship at a loss/support drop-shipping
2- merge maintenance buildings into stations
3- cut down track 'exclusivity' radius to permit stuff like parallel rails
4- add tool to import and export heightmaps
5- increase effect of slow delivery on perishable cargos
6- support express traffic to/from warehouses and ports
7- add new characters (not just portraits)
8- add cargo types
9- fix early exits from multiple-factor event loops
10- slightly increase possible zoom distance to cover huge maps

Harder:
1- add math expressions to events
2- add city-to-city or terr-to-terr load delivery tests
3- give player option to freeze AI after company 51% owned
4- fix event display in debug window for long scenarios
5- bring back train robberies from RT2
6- make it trivial to submit scenario reviews
7- add tool to import and export events
8- be able to delete territories
9- be able to turn off planes, non-evented storms, etc.
10- add cuts and fills when building rail through mountains

I can help out in the M, S and T areas, but would be willing and eager to learn F.

I'm really glad to see the high level of interest in modifying RT3 to get the things we want. It's kind of sad the SMR has dropped the ball on "improvements" to RT3 in favor of a "prettier" game.

Another feature that would be nice, but only of interest to the most masochistic of scenario developers, would be the ability to create campaigns. Although it would be a large task for an individual, a consortium of scenario writers could come up with campaigns worthy of play by the entire community. I know it's probably a low priority compared to the things already on the table, but would be really neat if it weren't too hard to do. It's probably just a matter of figuring out how several linked scenarios are connected.

[Gwizz]

I'm in agreement with most of the posts.

Milo, I will help where I can. My time is a bit pushed right now but I do take long breaks.

Outstanding project. Thanks for your effort.

PS: Is it possible to publish a patch to add to the game code, for a fee as long as you are not selling original code. Your efforts should be worth a return.

[milo]

Is it possible to publish a patch to add to the game code, for a fee as long as you are not selling original code. Your efforts should be worth a return.

NO. Something I gotta emphasize here, for everyone working on this project - this stuff stays free or it doesn't happen. Even though PopTop doesn't exist anymore, Take2 still owns the rights to the game, and I don't dare make it worth their while to come after us.

I'm also trying to keep as much of this open-source as possible, but that's more because there's no guarantee one of the developers won't be hit by a bus tomorrow than because of any legal requirement.

If you're worried about the return, think about this as a voluntary barter system. I like seeing new and complex RT3 maps; I'm not very good at constructing them, but I do have some skills at fixing the problems that have been blocking map-makers. Therefore, I'm contributing fixes and improvements merely in the hopes that it'll reawaken the interest that drove you to put together that nifty Roman Empire map.

[Good grief, getting Marxist in my old age...]

[kriss]

Milo,

You're the best thing since sliced bread!

I'm coming to the end of my other time consuming project now & Sid's latest doesn't grab me at all. Even my 6 year old grandson has abandoned it in favor of RRT3.

I'll be back to offer what help I can in a short time.

Keep up the fantastic work.